electric truck (BET) sales shares adjustments

[ahagen-pik]

Purpose of this PR

This PR addresses this issue: https://github.com/remindmodel/development_issues/issues/723

This PR adjusts electric truck sales shares to better reflect real-world near-term trends in the regions, special emphasis on:

• BET sales share is lowered in EUR, CAZ, JPN, NEU, USA in NPi2025 (esp. in 2030)
• BET sales share is strongly increased in CHA esp. in 2030 (from originally <25% to now ~55%)

This reflects a strong increase in BET sales shares seen in CHA in 2025 (as can be seen in the EV outlook 2026) and CHA as the region with highest BET sales shares globally.

For EUR the shares are still rather too high (now ~40% in 2030, NPi2025) compared with current shares and ACEA projections. This could possibly be further checked/validated and addressed in a subsequent step since a stronger adjustment needs a different methodology. Maybe the ICEban is too strong?

In addition:

• BET sales share is globally reduced in PkBudg750 (especially in 2030) to decrease the jump from Mix2 to Mix4 that was unplausible and present before in 2030
  This was also due to a unfavorable behavior of the function that is used to adjust the transportPolScen in EDGET from 2021 onward, and introduced a huge jump from 2020 to 2021 for some scenarios (esp. Mix4):

more context on that is in the issue linked above.

This was corrected in this PR by choosing different parameter sets for the function values for trucks, that prevent a huge jump.

However, the general behavior of the function is unchanged and setting new parameters needs still to be handled with care, and this problem may still be present for other modes or levels.

Checklist:

• I used ./test-standard-runs to compare and archive the changes introduced by this PR in /p/projects/edget/PRchangeLog/

Further information (optional):

• Test runs are here:

  • ChangeLog EDGE-T old vs. new: /p/projects/edget/PRchangeLog/20260529_PR405_adjBETsales/

  • REMIND runs with the adjustments included: /p/tmp/ahagen/remindRuns/202605_BETsales_remind_V04/
    (no new calibration)

• Comparison of results (what changes by this PR?):

  • compScen for REMIND runs NPi2025 & PKBudg750 before/after
    /p/projects/edget/PRchangeLog/20260529_PR405_adjBETsales/2026-05-29_10.26.03_BETsales_V04_REMINDH12.pdf

• cross comparison Mix1-4 EDGET standalone:
  /p/projects/edget/PRchangeLog/20260529_PR405_adjBETsales/2026-05-28_15.54.53_BETsales_V04H12.pdf

[ahagen-pik]

This also increases USA emissions mid-century a bit

https://github.com/remindmodel/development_issues/issues/740

[johannah-pik]

Looks good to me and more plausible! Thank you @ahagen-pik. (I haven't checked the data in detail, though. I assume you'll be cross-checking it with @robertpietzcker anyway)

[jmuessel]

Looks great, thank you @ahagen-pik!

One question regarding the US: Did the ES trajectories change? Because there are such changes in the sales fleet size. It's not strong but still significant. Independently, I think it's an improvement and does not change the story about BEV shares. I'll approve the PR now so that it makes it into the release. Thanks again for your work!